GAO-09-836 Reverse Mortgages: Policy Changes …NRMLA National Reverse Mortgage Lenders Association...

GAO United States Government Accountability Office

Report to Congressional Committees

REVERSE MORTGAGES

Policy Changes Have Had Mostly Positive Effects on Lenders and Borrowers, but These Changes and Market Developments Have Increased HUD’s Risk

July 2009

GAO-09-836

What GAO Found

United States Government Accountability Office

Why GAO Did This Study

HighlightsAccountability Integrity Reliability

July 2009 REVERSE MORTGAGES

Policy Changes Have Had Mostly Positive Effects on Lenders and Borrowers, but These Changes and Market Developments Have Increased HUD's Risk Highlights of GAO-09-836, a report to

congressional committees

Reverse mortgages—a type of loan against home equity available to seniors—are growing in popularity. A large majority of reverse mortgages are insured by the Department of Housing and Urban Development (HUD) under its Home Equity Conversion Mortgage (HECM) program. The Housing and Economic Recovery Act of 2008 (HERA) made several modifications to the HECM program, including changes in how origination fees are calculated and an increase in the loan limit. The Act directed GAO to examine (1) how these changes have affected lenders’ plans to offer reverse mortgages, (2) how the changes will affect borrowers, and (3) actions HUD has taken to evaluate the financial performance of the HECM program. To address these objectives, GAO surveyed a representative sample of HECM lenders, analyzed loan-level HECM data, and reviewed HUD estimates and analysis of HECM program costs.

What GAO Recommends

GAO makes no recommendations in this report. HUD concurred with the report’s findings.

On the basis of a survey of HECM lenders, GAO estimates that taken together, HERA’s changes to the HECM loan limit and origination fee calculation have had a positive to neutral influence on most lenders’ plans to offer HECMs. Other factors, such as economic and secondary market conditions, have had a mixed influence. Although economic conditions have had a positive influence on about half of lenders’ plans to offer HECMS, secondary market conditions have negatively influenced about one-third of lenders. GAO also estimates that the HERA changes have had little to no influence on most lenders’ plans to offer non-HECM reverse mortgages. HERA’s provisions will affect borrowers in varying ways depending on home value and other factors. The changes to HECM origination fees and loan limits are likely to change the up-front costs and the loan funds available for most new borrowers. GAO’s analysis of data on HECM borrowers from 2007 shows that if the HERA changes had been in place at the time, most would have paid less or the same amount in up-front costs, and most would have had more or the same amount of loan funds available. For example, about 46 percent of borrowers would have seen a decrease in up-front costs and an increase in available loan funds. However, 17 percent of borrowers would have seen an increase in up-front costs and a decrease in available loan funds. HUD has enhanced its analysis of HECM program costs, but less favorable house price trends and loan limit increases have increased HUD’s risk of losses. HUD has updated its cash flow model for the program and plans to conduct annual actuarial reviews. Although the program historically has not required a subsidy, HUD has estimated that HECMs made in 2010 will require a subsidy of $798 million, largely due to more pessimistic assumptions about long-run home prices. In addition, the higher loan limit enacted by HERA may increase the potential for losses. To calculate the amount of funds available to a borrower, lenders start with a limiting factor of either the home value or, if the home value is greater than the HECM loan limit, with the loan limit. For loans that are limited by the home value, the loan amount and the home value are closer together at the point of origination, which makes it more likely that the loan balance could exceed the home value at the end of the loan. In contrast, for loans that are limited by the HECM loan limit, there is initially a greater difference between the home value and the loan amount, making it less likely that the loan balance will exceed the home value at the end of the loan. The increase in the HECM loan limit may increase HUD’s risk of losses by reducing the proportion of loans that are limited by the HECM loan limit.

View GAO-09-836 or key components. For more information, contact Mathew J. Scirè at (202) 512-8678 or [email protected].

http://www.gao.gov/products/GAO-09-836

mailto:[email protected]

http://www.gao.gov/cgi-bin/getrpt?GAO-09-836

Page i GAO-09-836

Contents

Letter 1

Background 3 Most HECM Lenders View the Overall Effect of the HERA

Provisions as Neutral or Positive for Their Reverse Mortgage Business 11

HERA Provisions Will Affect Borrower Costs and Loan Amounts Differently Depending on Home Value and Other Factors 20

HUD Has Enhanced Its Analysis of HECM Program Costs but Changes in House Price Trends and Higher Loan Limits Have Increased HUD’s Risk of Losses 27

Agency Comments and Our Evaluation 37

Appendix I Objectives, Scope, and Methodology 38

Appendix II Impact of Loan Limit Increase in the American

Recovery and Reinvestment Act of 2009 on HECM

Lenders 45

Appendix III Effect of the Housing and Economic Recovery Act of

2008 on Up-front Costs for HECM Borrowers 46

Appendix IV GAO Contact and Staff Acknowledgments 48

Tables

Table 1: Change in Up-front Costs for $300,000 Houses in Various Locations 21

Table 2: Survey Population and Sample Dispositions 39 Table 3: Universe of 2007 HECMs by Home Value and Maximum

Claim Amount Category 43

Figures

Figure 1: Comparison of 30-Year Forward and Reverse Mortgages 4

Reverse Mortgages

Figure 2: Number of HECMs Insured Annually, Fiscal Years 1990 through 2008 5

Figure 3: Potential Liability of Active HECM Loans, Fiscal Years 2006 through 2008 6

Figure 4: Number of HUD-Approved Lenders Originating HECMs, Fiscal Years 2003 through 2008 7

Figure 5: FHA Process to Determine the Maximum Claim Amount and the Amount of Loan Funds Available 9

Figure 6: Influence of HERA’s Provisions on Loan Limits and Fees and Other Factors on Lenders’ Plans to Offer HECMs 13

Figure 7: Influence of HERA’s Provisions on Loan Limits and Fees and Other Factors on Lenders’ Plans to Offer Non-HECM Reverse Mortgages in 2009 19

Figure 8: Number of 2007 HECM Borrowers Affected by HERA Provisions 24

Figure 9: Average Changes in Maximum Claim Amounts, Up-front Costs, and Loan Funds Available for 2007 HECM Borrowers 25

Figure 10: HECM Credit Subsidy Rates, Fiscal Years 2006 through 2010 31

Figure 11: Loan Liability Guarantee for the HECM Program, Fiscal Years 2006 through 2008 32

Figure 12: Illustration of How Increasing the HECM Loan Limit Could Increase HUD’s Losses 34

Figure 13: Percentage of HECMs with Maximum Claim Amounts Limited by the Program Limit 36

Figure 14: Influence of ARRA’s Increase to Loan Limits on Lenders’ Plans to Offer Reverse Mortgages 45

Figure 15: Influence of ARRA’s Increase to Loan Limits on Consumer Demand for HECMs 45

Figure 16: Changes to Up-front Costs for Borrowers Not Affected by HERA’s Change in Loan Limit 47

Figure 17: Changes to Up-front Costs for Borrowers Affected by HERA’s Change in Loan Limit 47

Page ii GAO-09-836 Reverse Mortgages

Abbreviations

ARRA American Recovery and Reinvestment Act of 2009 CBO Congressional Budget Office FCRA Federal Credit Reform Act FHA Federal Housing Administration GI/SRI General Insurance and Special Risk Insurance HECM Home Equity Conversion Mortgage HERA Housing and Economic Recovery Act of 2008 HMBS HECM Mortgage Backed Security HUD Department of Housing and Urban Development LLG liability for loan guarantees MBA Mortgage Bankers Association MMI Mutual Mortgage Insurance NRMLA National Reverse Mortgage Lenders Association OIG Office of the Inspector General

This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately.

Page iii GAO-09-836 Reverse Mortgages

GAO-09-836

he

United States Government Accountability Office

Washington, DC 20548

July 30, 2009

The Honorable Christopher Dodd Chairman The Honorable Richard Shelby Ranking Member Committee on Banking, Housing, and Urban Affairs United States Senate The Honorable Barney Frank Chairman The Honorable Spencer Bachus Ranking Member Committee on Financial Services House of Representatives

A reverse mortgage is a loan that converts the borrower’s home equity into payments from a lender and typically does not require any repayment as long as the borrower continues to live in the home. Available to homeowners aged 62 and older, these loans have become an increasingly popular financial tool for seniors. Almost all reverse mortgages are currently made under the Home Equity Conversion Mortgage (HECM) program administered by the Federal Housing Administration (FHA) of the Department of Housing and Urban Development (HUD).1 FHA insures lenders against losses on these mortgages, and charges borrowers insurance premiums to cover anticipated insurance claims. The number of HECMs made has grown rapidly in recent years, rising from 157 loans in fiscal year 1990 to more than 112,000 loans in fiscal year 2008.

The Housing and Economic Recovery Act of 2008 (HERA) made several modifications to the HECM program.2 The first of these changes affects the origination fees that borrowers pay for these loans. Prior to HERA, torigination fee was 2 percent of the “maximum claim amount”—the lesser of the home value or the HECM loan limit. HERA changed the fee calculation to 2 percent of the maximum claim amount up to $200,000 plus

1For information on consumer protection issues regarding HECMs, see GAO, Product

Complexity and Consumer Protection Issues Underscore Need for Improved Controls

over Counseling for Borrowers, GAO-09-606 (Washington, D.C.: June 29, 2009).

2P.L. 110-289

Reverse Mortgages

http://www.gao.gov/cgi-bin/getrpt?GAO-09-606

1 percent of the maximum claim amount over $200,000 with a maximum fee of $6,000. In conjunction with this change, HUD increased the minimum origination fee from $2,000 to $2,500 to ensure that lenders retain an incentive to continue to serve borrowers living in lower-valued properties.3 The second of these changes affects the program’s loan limit, the maximum loan amount that HUD can insure. Specifically, HERA established a national loan limit for HECMs, which was set at $417,000—a level substantially higher than the county-based limits that existed prior to HERA.4

In light of these changes, HERA contains a mandate for GAO to evaluate the impact of HERA’s provisions on the availability of credit under the HECM program, the cost to borrowers participating in the program, and the program’s financial soundness. As agreed with your offices, this report examines (1) how HERA’s changes to the HECM program and other factors have affected HECM lenders’ planned participation in the reverse mortgage market, (2) the extent to which HERA’s changes to HECM origination fees and loan limits will affect costs to borrowers and the loan amounts available to them, and (3) HUD’s actions to evaluate the financial performance of the HECM program, including the potential impact of loan limit and house price changes.

To address these objectives, we reviewed laws, regulations, and guidance relevant to the HECM program, including provisions in HERA and the American Recovery and Reinvestment Act of 2009 (ARRA). We also spoke with agency, industry, and nonprofit officials, including those at HUD, Ginnie Mae, Fannie Mae, AARP, the National Reverse Mortgage Lenders Association (NRMLA), and the Mortgage Bankers Association (MBA) about the implications of the HERA changes. In addition, to determine how HERA’s provisions have affected lenders’ planned participation in the reverse mortgage market we conducted a survey of a representative sample of lenders that originated 10 or more HECMs in fiscal year 2008. The survey included questions about lenders’ plans to offer reverse mortgages, resources dedicated to their HECM business, consumer

3The Secretary of HUD has the authority to set a minimum origination fee. For purposes of this report, our use of the terms “HERA provisions” and “HERA changes” includes HUD’s change to the program’s minimum origination fee.

4Prior to HERA, some parts of Hawaii had HECM loan limits exceeding $417,000. HERA did not change those limits. As a result of the American Recovery and Reinvestment Act of 2009, the HECM loan limit was increased in all areas of the country to $625,500 through December 31, 2009.

Page 2 GAO-09-836 Reverse Mortgages

demand for HECMs, and margin rates for HECMs. We received responses from 57 percent of the lenders we surveyed. In general, the estimates we made from our survey results had margins of error of plus or minus 10 percentage points at the 95 percent confidence interval. To determine the extent to which HERA’s changes to HECM origination fees and loan limits have affected costs to borrowers and the loan amounts available to them, we reviewed rules for determining borrower costs and loan amounts. We also obtained and analyzed loan-level data from HUD on HECM loans and borrowers, which we determined to be reliable for the purposes of this report. We compared the actual up-front costs and loan funds available to borrowers who obtained HECMs in 2007 to what their costs and available loan funds would have been under the HERA provisions. To examine HUD’s actions to evaluate the financial performance of the HECM program, we reviewed HUD budget estimates, financial statements, and actuarial reviews for the HECM program, as well as other analyses the agency has conducted of program costs. We also reviewed HUD Office of the Inspector General (OIG) audits of FHA’s financial statements. Additionally, we reviewed federal agency standards for managing credit programs, such as those contained in the Federal Credit Reform Act (FCRA), related Office of Management and Budget requirements and instructions, and Federal Accounting Standards Advisory Board guidance. Finally, we interviewed FHA officials, HUD OIG officials, industry participants, and mortgage market analysts.

We conducted this performance audit from September 2008 through July 2009, in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. A more extensive discussion of our scope and methodology appears in appendix I.

A reverse mortgage is a loan against the borrower’s home that the borrower does not need to repay for as long as the borrower meets certain conditions. These conditions, among others, require that borrowers live in the home, pay property taxes and homeowners’ insurance, maintain the property, and retain the title in the borrower’s name. Reverse mortgages typically are “rising debt, falling equity” loans, in which the loan balance increases and the home equity decreases over time. As the borrower receives payments from the lender, the lender adds the principal and interest to the loan balance, reducing the homeowner’s equity. This is the

Background


opposite of what happens in forward mortgages, which are characterized as “falling debt, rising equity” loans. With forward mortgages, monthly loan payments made to the lender add to the borrower’s home equity and decrease the loan balance (see fig. 1).

Figure 1: Comparison of 30-Year Forward and Reverse Mortgages

0

100

200

300

400

500

600

302928272625242322212019181716151413121110987654321

0

100

200

300

400

500

600

302928272625242322212019181716151413121110987654321

Dollars (in thousands)


Age of loan in years


Traditional 30-year forward mortgage

Reverse mortgage, over 30 years

Source: GAO.

Home equity

Loan balance (debt)

Note: The graphs are based on a starting house value of $300,000, with an annual 2 percent home appreciation rate. The interest rates are assumed to be fixed at 5 percent for the forward mortgage and 3 percent for the reverse mortgage.

There are two primary types of reverse mortgages, HECMs and proprietary reverse mortgages. The Housing and Community Development Act of 1987 (P.L. 100-242) authorized HUD to insure reverse mortgages and established the HECM program. According to industry officials, HECMs account for more than 90 percent of the market for reverse mortgages. Homeowners aged 62 or older with a significant amount of home equity are eligible, as long as they live in the house as the principal residence, are not delinquent on any federal debt, and live in a single-family residence. If


the borrower has any remaining balance on a forward mortgage, this generally must be paid off first (typically, taken up-front from the reverse mortgage). In addition, the condition of the house must meet HUD’s minimum property standards, but a portion of the HECM can be set aside for required repairs. The borrower makes no monthly payments, and there are no income or credit requirements to qualify for the mortgage. Lenders have offered non-HECM, or proprietary, reverse mortgages in the past, but these products have largely disappeared from the marketplace due, in part, to the lack of a secondary market for these mortgages. Typically, proprietary reverse mortgages have had higher loan limits than HECMs but paid out a lower percentage of the home value to borrowers.

The volume of HECMs made annually has grown from 157 loans in fiscal year 1990 to more than 112,000 loans in fiscal year 2008. The HECM program has experienced substantial growth, as the number of HECMs insured by FHA has nearly tripled since 2005 (see fig. 2).

Figure 2: Number of HECMs Insured Annually, Fiscal Years 1990 through 2008

0

20

40

60

80

100

120

2008200720062005200420032002200120001999199819971996199519941993199219911990

HECMs (in thousands)

Source: GAO analysis of HECM data.

Fiscal year

Additionally, the potential liability of loans insured by FHA has doubled in the last 2 years (see fig. 3). The potential liability is the sum of the maximum claim amounts for all active HECMs since the program’s inception.


Figure 3: Potential Liability of Active HECM Loans, Fiscal Years 2006 through 2008

0

10

20

30

40

50

60

70

80

90

200820072006

Dollars (in billions)

Source: GAO analysis of FHA’s financial statements (2006 through 2008).

Fiscal year

Finally, recent years have seen a rapid increase in the number of lenders participating in the HECM program (see fig. 4). However, the bulk of HECM business is concentrated among a relatively small percentage of lenders. In fiscal year 2008, roughly 80 percent of all HECMs were originated by fewer than 300 lenders, or about 10 percent of HECM lenders.


Figure 4: Number of HUD-Approved Lenders Originating HECMs, Fiscal Years 2003 through 2008

0

500

1,000

1,500

2,000

2,500

3,000

200820072006200520042003

Lenders

Source: GAO analysis of HECM data.

Fiscal year

Lenders can participate in the HECM market through wholesale or retail channels. Wholesale lenders fund loans originated by other entities, including mortgage brokers and loan correspondents. Retail lenders originate, underwrite, and close loans without reliance on brokers or loan correspondents. Most lenders participate in the HECM market through retail lending, although some participate through the wholesale process, and a few have both a retail and wholesale HECM business.

There is a secondary market for HECMs, as most lenders prefer not to hold the loans on their balance sheets. Fannie Mae has purchased 90 percent of HECM loans and holds them in its portfolio. In 2007, Ginnie Mae developed and implemented a HECM Mortgage Backed Security product, in which Ginnie Mae-approved issuers pool and securitize a small proportion of HECMs. Fannie Mae and Ginnie Mae’s involvement in the HECM secondary market helps to provide liquidity so that lenders can continue offering HECM loans to seniors.

The amount of loan funds available to the borrower is determined by several factors (see fig. 5).


• First, the loan amount is based on the “maximum claim amount,” which is the highest sum that HUD will pay to a lender for an insurance claim on a particular property. It is determined by the lesser of the appraised home value or the HECM loan limit. In the past year, Congress has raised the HUD loan limit for HECMs twice: HERA established for the first time a national limit for HECMs, which was set at $417,000. As a result of ARRA, the national limit was raised again to $625,500 through December 31, 2009. Prior to HERA, the loan limit for HECMs varied by location and generally were set at 95 percent of the local area median house price.

• Second, to manage its insurance risk, HUD limits the loan funds available to the borrower by applying a “principal limit factor” to the maximum claim amount.5 HUD developed a principal limit factor table using assumptions about loan termination rates—which are influenced by borrower mortality and move-out rates—and long-term house price appreciation rates, and indexed the table by (1) the borrower’s age and (2) the expected interest rate—the 10-Year Treasury rate plus the lender’s margin. The lender determines which factor to use by inputting the borrower’s current age and the current interest rate information. The older the borrower, the higher the loan amount; the greater the expected interest rate of the loan, the smaller the loan amount.

• Third, the funds available to the borrower are further reduced by a required servicing fee set-aside and by the up-front costs (which include a mortgage insurance premium and the origination fee), because borrowers can choose to finance them. HUD allows lenders to charge up to $35 as a monthly HECM servicing fee. The lender calculates the servicing fee set-aside by determining the total net present value of the monthly charged servicing fees that the borrower would pay between loan origination and when the borrower reaches age 100.6 The set-aside limits the loan funds available but is not added to the loan balance at origination. If borrowers choose to finance up-front costs as part of the loan, the loan funds available are reduced by these costs.

5The principal limit factor can range from 20.4 to 90 percent of the maximum claim amount.

6Present value expresses the worth a future stream of cash inflows and outflows in terms of an equivalent lump sum received (or paid) today. Net present value is the present value of estimated future cash inflows minus the present value of estimated future cash outflows.


Figure 5: FHA Process to Determine the Maximum Claim Amount and the Amount of Loan Funds Available

Source: GAO.

Up-front mortgageinsurance premium

+ origination fee

Maximumclaim amount

Principallimit

factor

Up-frontcosts

Servicingfee set aside

Loanamount

Loan fundsavailable

X = + =_

FHA loanlimit

Housevalue

Lesservalue of or

Borrower’sage

10 yr.Treasury

rate +lendermargin

Borrowers incur various costs when obtaining a HECM. HUD allows borrowers to finance both up-front and long-term costs through the loan, which means they are added to the loan balance.

• Origination fee: Prior to HERA, HECM borrowers were charged an origination fee equal to 2 percent of the maximum claim amount with a minimum fee of $2,000. Since the implementation of HERA, HECM borrowers are charged an origination fee calculated as 2 percent of the maximum claim amount up to $200,000 plus 1 percent of the maximum claim amount over $200,000, with a maximum fee of $6,000 and a minimum fee of $2,500.

• Mortgage insurance premium: Borrowers are charged an up-front mortgage insurance premium equal to 2 percent of the maximum claim amount. While the maximum claim amount is always higher than the initial amount a borrower can receive in HECM payments from the lender, FHA charges the mortgage insurance premium based on this amount because the loan balance (with accumulated interest and fees) could exceed the amount a borrower receives in payments and potentially reach the maximum claim amount. Additionally, borrowers are charged a monthly mortgage insurance premium on their loan balance at an annual rate of 0.5 percent.

• Interest: Borrowers are charged interest, which generally includes a base interest rate plus a fixed lender margin rate, on the loan balance. Lenders can offer HECMs with fixed, annually adjustable, or monthly adjustable base interest rates. The adjustable rates can be tied to either the 1-Year


Constant Maturity Treasury Rate or 1-Year London Interbank Offered Rate Index. Most HECMs have adjustable interest rates.

• HECM counseling fee: The HECM program requires prospective borrowers to receive counseling to ensure an understanding of the loan. HUD allows counseling providers to charge borrowers up to $125 for HECM counseling.

• Loan servicing fee: Borrowers pay a monthly servicing fee of up to $35.

• Closing costs: HECMs also have other up-front closing costs, such as appraisal and title search fees.

FHA’s insurance for HECMs protects borrowers and lenders in four ways. First, lenders can provide borrowers with higher loan amounts than they could without the insurance. Second, when the borrower is required to repay the loan to the lender, if the proceeds from the sale of the home do not cover the loan balance, FHA will pay the lender the difference. Third, if the lender is unable to make payments to the borrower, FHA will assume responsibility for making these payments. Fourth, if the loan balance reaches 98 percent of the maximum claim amount, the lender may assign the loan to FHA and FHA will continue making payments to the borrower if the borrower has remaining funds in a line of credit or still is receiving monthly payments. To cover expected insurance claims, FHA charges borrowers insurance premiums, which go into an insurance fund. HECM loans originated since the inception of the program through 2008 are supported by FHA’s General Insurance and Special Risk Insurance Fund, which includes a number of FHA mortgage insurance programs for single-family and multifamily housing and hospitals. Pursuant to HERA, FHA moved the HECM program and other insurance programs for single-family housing into FHA’s Mutual Mortgage Insurance Fund.

FCRA requires federal agencies that provide loan guarantees to estimate the expected cost of programs by estimating their future performance and reporting the costs to the government in their annual budgets.7 Under credit reform procedures, the cost of loan guarantees, such as mortgage insurance, is the net present value of all expected cash flows, excluding administrative costs. This is known as the credit subsidy cost. For loan guarantees, cash inflows consist primarily of fees and premiums charged

7FCRA was enacted as part of the Omnibus Budget Reconciliation Act of 1990 (P.L. 101-508).


to insured borrowers and recoveries on assets, and cash outflows consist mostly of payments to lenders to cover the cost of claims. Annually, agencies estimate credit subsidy costs by cohort, or all the loans the agency is committing to guarantee in a given fiscal year. The credit subsidy cost can be expressed as a rate. For example, if an agency commits to guarantee loans totaling $1 million and has estimated that the present value of cash outflows will exceed the present value of cash inflows by $15,000, the estimated credit subsidy rate is 1.5 percent. When estimated cash inflows exceed estimated cash outflows, the program is said to have a negative credit subsidy rate. When estimated cash outflows exceed estimated cash inflows, the program is said to have a positive credit subsidy rate—and therefore requires appropriations.

Generally, agencies are required to produce annual updates of their subsidy estimates—known as re-estimates—of each cohort based on information about the actual performance and estimated changes in future loan performance. This requirement reflects the fact that estimates of subsidy costs can change over time. Beyond changes in estimation methodology, each additional year provides more historical data on loan performance that may influence estimates of the amount and timing of future claims. Economic assumptions also can change from one year to the next, including assumptions on home prices and interest rates. FCRA recognized the difficulty of making subsidy cost estimates that mirrored actual loan performance and provides permanent and indefinite budget authority for re-estimates that reflect increased program costs.

In combination, HERA’s changes to the HECM loan limit and origination fee calculation have had a positive to neutral influence on most lenders’ plans to start or continue offering HECMs. Other factors have had varying influences on lenders’ planned participation. Current economic conditions have had a moderate upward influence on lenders’ plans; however, secondary market conditions have had a downward influence on about one-third of lenders’ plans to start or continue offering HECMs. Finally, the HERA changes have not influenced most lenders’ plans to offer proprietary—non-HECM—products.

Most HECM Lenders View the Overall Effect of the HERA Provisions as Neutral or Positive for Their Reverse Mortgage Business


HERA’s changes to the HECM program have had varying effects on HECM lenders’ planned participation in the HECM market. On the basis of questionnaire responses from a random sample of HECM lenders, we estimate that for 50 percent of lenders, the combined effect of these changes has had an upward influence on their plans to start or continue to offer HECMs (see fig. 6).8 For 42 percent of lenders, the combination of HERA’s changes to the origination fee and loan limits for the HECM program have had little to no influence on their plans to offer HECMs, while for 8 percent of lenders, HERA’s changes have had a downward influence. Some industry participants we interviewed stated that the changes were a good compromise that benefited borrowers by limiting the origination fee and increasing the loan limit, thereby increasing the money borrowers could receive from a HECM. Additionally, officials at NRMLA and MBA said the changes benefited lenders by making the product more attractive to individuals with higher-value homes.

HERA’s Changes and Other Factors Have Had Varying Effects on Lenders’ Planned Participation in the HECM Market

8We surveyed a random sample of the 2,779 lenders that originated HECMs on a retail basis in fiscal year 2008. For purposes of this report, we define retail lenders as lenders that originate HECMs as opposed to funding HECMs originated by other lenders. For our survey questions about HERA’s changes and other factors influencing lenders’ planned participation, we asked lenders the following: “How, if at all, has (factor x) influenced your institution’s likelihood to start or continue to offer HECMs on a retail basis?” Accordingly, our results apply only to lenders’ retail HECM business. Unless otherwise noted, our estimates have margins of error of plus or minus 10 percentage points or less at the 95 percent confidence interval. See appendix I for additional information on this survey methodology.


Figure 6: Influence of HERA’s Provisions on Loan Limits and Fees and Other Factors on Lenders’ Plans to Offer HECMs

Current secondary market options

Current availability ofwholesale lending partners

HERA's prohibition oflender-funded HECM counseling

Implementation of theHECM for Purchase program

HERA's prohibition of participation of non-FHA approved entities in the origination of HECMs

HERA's restrictions on sellingother financial products in conjunction with HECMs

Reduced opportunities in theforward mortgage market

Current economic conditions

Current house price trends

HERA's changes to both HECMorigination fees and loan limits

HERA's changes toHECM loan limits

HERA's changes to thecalculation of origination fees 11

70

50

67

52

34

33

14

10

9

9

5

% 65

28

42

29

28

54

48

50

50

60

56

78

% 22

1

8

1

19

8

13

32

38

29

22

7

% 2

1

1

2

0

0

4

0

1

1

1

6

% N/A

N/A

N/A

0

0

4

1

4

1

0

12

4

%

Don’tknow

Great to moderatedownward influence

Little or noinfluence

Moderate or greatupward influence

Notapplicable

Source: GAO analysis of survey of HECM lenders.

HERA’s provisions on loanlimits and fees

Other factors

Note: Figure shows estimates based on GAO survey of HECM lenders. Estimates have margins of error of plus or minus 10 percentage points or less at the 95 percent confidence interval.

Taken separately, the two HERA provisions have had differing effects on lenders’ plans to offer HECMs. We estimate that for about 70 percent of lenders, HERA’s increase in HECM loan limits has had an upward


influence on the likelihood of offering HECMs.9 The loan limit increase has had little to no influence on almost all of the remaining lenders’ plans to offer HECMs. We estimate that 86 percent of lenders expect that HERA’s creation of a single national loan limit of $417,000 will somewhat or greatly increase consumer demand for HECMs.

Although the increase in the loan limit has generally had an upward influence on lenders’ plans, the change to the calculation of the origination fee has had a different effect. We estimate that changing how the fee is calculated has had a downward influence on plans to offer HECMs for 22 percent of HECM lenders, little to no influence for 65 percent of lenders, and an upward influence for 11 percent of lenders. Consistent with these views, 65 percent of lenders expect the change in origination fee to have no effect on consumer demand for HECMs. An estimated 26 percent of lenders expect the change in the origination fee to increase consumer demand, while only a few lenders expect the change to decrease consumer demand.

We estimate that only 2 percent of HECM lenders do not plan to continue to offer HECMs. Of the respondents in our sample, three lenders indicated that they did not plan to continue offering HECMs. None of these were large HECM lenders, as they each originated from 40 to 160 HECMs in fiscal year 2008. Each of these lenders participated in the HECM market solely through their retail business. These three lenders varied in the amount of time that they have offered the HECM product. A representative of one lender indicated that HERA’s changes to the loan limits and origination fee had a great upward influence on the likelihood that it would offer HECMs, but nonetheless planned to discontinue offering HECMs. The other two lenders indicated that HERA and other economic factors had little to no influence on their decision to discontinue offering HECMs, and one of these lenders noted on the survey that it had discontinued offering HECMs before the enactment of the HERA.

As part of our survey, we asked lenders how various economic and legislative factors influenced their plans to start or continue offering HECMs. Two factors had an upward influence on most lenders’ plans to offer HECMs in 2009. For an estimated 67 percent of HECM lenders, the implementation of the HECM for Purchase program (authorized by HERA)

9See appendix II for survey results pertaining to the ARRA’s increase in the HECM loan limit to $625,500.


has had an upward influence on their plans to offer HECMs, and it has had little to no influence on almost all of the remaining lenders’ HECM origination plans.10 Some industry participants told us that the HECM for Purchase program likely will make HECMs attractive to a broader range of seniors. Additionally, current economic conditions have had an upward influence on the plans to offer HECMs for about 52 percent of lenders. NRMLA officials explained that seniors are seeking additional revenue because they have less available income from traditional sources, such as interest and dividend payments and retirement accounts, which is partially attributable to poor economic and financial market conditions. Additionally, two other factors have had an upward influence on some lenders’ plans to offer HECMS. For about one-third of lenders, both (1) reduced opportunities in the forward mortgage market and (2) HERA’s prohibition on the participation of non-FHA approved entities in the origination of HECMs has had a moderate or great upward influence on their plans to offer HECMs.11

In contrast, three factors had more of a downward influence on some lenders’ planned participation in the HECM market. First, we estimate from our survey that house price trends have had a downward influence on the HECM origination plans of 38 percent of lenders; however, house price trends had little or no influence on plans for about 50 percent of lenders. Some industry participants told us that the recent decline in house prices has prevented some seniors from obtaining a HECM either because they lack the equity in their home to qualify for the loan, or because they would not receive enough funds from the HECM to have any cash remaining after they deduct HECM fees and pay off any existing mortgage debt.

Second, we estimate that the availability of secondary market options has had a downward influence on the plans of about one-third of lenders to offer HECMs. The secondary market for HECMs plays an important role in

10HERA authorized a HECM for Purchase program for seniors who wish to use a HECM to buy a new home. Unlike a traditional HECM, a HECM for purchase is made against the value of the home to be purchased rather than against the value of a home the borrower already owns. The HECM for purchase program allows a senior to simultaneously buy a new home and obtain a HECM in a single transaction with a single set of closing costs, reducing the cost to the senior.

11HERA requires that all parties participating in the origination of HECMs be approved by HUD. This prohibits non-HUD-approved mortgage brokers—sometimes called HECM advisors—from offering HECMs.


maintaining availability of loans because lenders prefer not to hold HECMs on their balance sheets. There are currently two primary options in the secondary market—Fannie Mae and Ginnie Mae.

• Fannie Mae officials stated that Fannie Mae bought and held more than 90 percent of HECMs in its portfolio in 2008 and was the principal secondary market purchaser of HECM loans. However, Fannie Mae’s regulator—the Federal Housing Finance Agency—recently required it to reduce the mortgage assets it holds in portfolio. Fannie Mae officials told us that as a result, they are making changes to their HECM business, which will attract other investors to the secondary market for HECMs, in order to decrease their share of the market. Recently, Fannie Mae lowered the price it pays lenders for HECMs and implemented a “live pricing” system that requires lenders to commit to the volume of HECMs they will sell to Fannie Mae.12 We estimate that approximately 90 percent of lenders viewed secondary market pricing requirements and the transition to live pricing as important factors in recent margin rate increases on HECMs. Fannie officials explained that as the price they pay lenders for HECMs falls, the margin rate the lenders charge the consumers generally increases.13 Some lenders we surveyed noted that margin rate increases stemming from pricing changes could make HECMs less attractive to borrowers because they would not be able to obtain as much cash from their HECM.14 Some lenders noted that live pricing complicates their relationship with borrowers because the interest rate can change between loan application and closing, which may result in the senior being able to receive less money from their HECM than originally quoted.

12Live pricing eliminated the predetermined pricing contracts that were previously used, in which lenders negotiated with Fannie Mae to set the price of all loans made within a predetermined period. Instead, live pricing implemented a process whereby lenders enter into a commitment with Fannie Mae to sell a specified volume of loans at a specified price within 90 days. Fannie Mae officials noted that this system makes HECM pricing more similar to the pricing system for forward mortgages in the secondary market. According to some industry observers, some lenders are reluctant to guarantee the interest rate on a HECM well in advance of closing because it can take several months to close a HECM and the loan may not close before the end of the 90-day commitment period, in which case the lender would incur a penalty.

13Fannie Mae maintains a cap on margin rates for HECMs they purchase.

14As previously discussed, the amount of loan funds available to the borrower depends in part on the interest rate, which includes the lender’s margin. In general, the higher the interest rate, the less money a borrower will have available from the HECM.


• Ginnie Mae developed and guarantees a HECM Mortgage Backed Security (HMBS) that aims to expand the availability of HECMs from multiple lenders, reduce borrowing costs, and create a broader secondary market for HECM loans. Ginnie Mae officials stated that they were poised to take on extra volume in the HECM secondary market by guaranteeing securities issued by lenders. AARP officials noted that Ginnie Mae’s HMBS product could help introduce competition into the secondary market for reverse mortgages, lowering margin rates for seniors. However, industry participants point to several issues with the Ginnie Mae product that could limit its appeal to lenders. First, Ginnie Mae requires HMBS issuers to buy back the HECM when the loan balance reaches 98 percent of the loan’s maximum claim amount.15 Second, issuers are required to pay interest shortfalls to investors when the loan is terminated mid-month. Some HECM lenders have noted that both of these provisions expose them to extra risk on the loan, as compared to the alternative of selling the HECM outright as they had when selling to Fannie Mae.

Third, for an estimated 29 percent of lenders, HERA’s prohibition on lender-funded counseling has had a downward influence on plans to offer HECMs. Industry participants said that this prohibition is a problem for the HECM industry because counseling is required for borrowers to obtain a HECM, but borrower-paid counseling can be a deterrent for seniors who are still deciding if they want a HECM, or for those who have limited financial means to pay for counseling. In contrast to these comments, we estimate that the prohibition on lender-funded counseling had little or no influence on the plans of 60 percent of lenders.

Our survey of HECM lenders asked about two other factors—HERA’s restrictions on selling other financial products in conjunction with HECMs and the current availability of wholesale lending partners—that could influence lenders’ plans to start or continue to offer HECMs. In general, these factors had little or no influence on lenders’ plans (see fig. 6).

HERA Has Not Influenced Most Lenders’ Plans to Offer Non-HECM Reverse Mortgages

In 2008, several non-HECM reverse mortgages—referred to as jumbo or proprietary reverse mortgages—were available in the marketplace. Proprietary reverse mortgages offered loan limits that were greater than the HECM loan limit. For example, Financial Freedom, a large reverse mortgage lender, offered a product called the Cash Account Advantage Plan, which was not subject to the HECM loan limits, and in some cases

15Generally, issuers can then assign the loan to HUD.


provided more cash than a HECM to borrowers with higher-value homes. Based on our survey results, we estimate that approximately 43 percent of HECM lenders made non-HECM reverse mortgages in 2008. However, towards the end of 2008, almost all of the non-HECM reverse mortgage products were withdrawn from the market due to the lack of a secondary market to support them. Nonetheless, from our survey results, we estimate that 36 percent of HECM lenders plan to offer a non-HECM reverse mortgage in 2009.

We estimate that HERA’s changes to the calculation of the origination fee and loan limit have had little or no influence on 68 percent of lenders’ plans to originate non-HECM reverse mortgages (see fig. 7). However, for an estimated 29 percent of HECM lenders, HERA’s change to the loan limits has had an upward influence on their plans to offer non-HECM reverse mortgages. Additionally, we estimate that for 32 percent of lenders, the implementation of the HECM for Purchase program had an upward influence on their plans to offer these loans. We estimate that current economic conditions have had an upward influence on plans to offer non-HECM reverse mortgages for 29 percent of lenders, little to no influence for 34 percent of lenders, and a downward influence for 17 percent of lenders. Our survey of HECM lenders asked about several other factors (see fig. 7) that could influence lenders’ plans to offer a non-HECM reverse mortgage product in 2009. Generally, these factors have had little or no influence on lenders’ plans. Our survey results did not indicate that secondary market conditions had a downward influence on the plans of most lenders. However, several lenders we interviewed said that while they hoped to offer a non-HECM reverse mortgage in 2009, their ability to do so would depend on the availability of funding in the secondary market.


Figure 7: Influence of HERA’s Provisions on Loan Limits and Fees and Other Factors on Lenders’ Plans to Offer Non-HECM Reverse Mortgages in 2009

Current secondary market options

Current availability ofwholesale lending partners

HERA's prohibition oflender-funded HECM counseling

Implementation of theHECM for Purchase program

HERA's prohibition of participation of non-FHA approved entities in the origination of HECMs

HERA's restrictions on sellingother financial products in conjunction with HECMs

Reduced opportunities in theforward mortgage market

Current economic conditions

Current house price trends

HERA's changes to both HECMorigination fees and loan limits

HERA's changes toHECM loan limits

HERA's changes to thecalculation of origination fees 9

29

15

32

29

19

17

16

14

12

8

5

% 74

56

68

43

34

38

53

57

45

45

61

71

% 12

10

12

5

17

22

7

6

21

24

12

3

% 5

6

5

4

3

5

4

5

3

3

4

6

% N/A

N/A

N/A

16

16

17

19

17

16

16

16

15

%

Don’tknow




Notapplicable


HERA’s provisions on loanlimits and fees

Other factors



HERA’s provisions will affect borrowers in varying ways depending primarily on home value and whether HERA’s increase in loan limit will change the maximum claim amount of the loan. HERA’s changes to HECM origination fees and loan limits are likely to change the up-front costs (origination fee and up-front mortgage insurance premium) and the loan funds available for most new borrowers. Our analysis of data on borrowers who took out HECMs in 2007 shows that had the HERA provisions been in place, most borrowers would have paid less or the same amount in up-front costs, and most would have had more or the same amount of loan funds available. Additionally, about 28 percent of HECM borrowers in 2007 would have seen an increase in maximum claim amount due to HERA’s increase in loan limit, which would have meant more loan funds available for nearly all of these borrowers. Borrowers also may be affected by other consequences of the HERA provisions, such as margin rate increases and changes to funding of HECM counseling.

HERA Provisions Will Affect Borrower Costs and Loan Amounts Differently Depending on Home Value and Other Factors

HERA Provisions Will Change Up-front Costs for Many Borrowers

The net effect of the HERA provisions on an individual borrower’s total up-front costs depends on house value, the local loan limit prior to HERA, and the new loan limit. HECM up-front costs consist primarily of the up-front mortgage insurance premium and the origination fee, both of which are calculated as a proportion of the maximum claim amount. Most borrowers are likely to see changes in origination fees due to HERA. Generally, those with house values greater than the prior HECM loan limit in their area will see changes in the up-front mortgage insurance premium.16 Borrowers fall into two categories, based on whether their maximum claim amount changes:

• Maximum claim amount does not change: For borrowers whose houses are valued at or less than the prior HECM loan limit in their area, the maximum claim amount does not change. Therefore, for these borrowers, the mortgage insurance premium (which is calculated based on the maximum claim amount) also does not change. However, the origination fee may change depending on the value of the house. A borrower whose house is valued at less than $125,000 should expect up to a $500 increase in the up-front costs due to the increase in the minimum origination fee from $2,000 to $2,500. A borrower whose house is valued at $125,000 to

16HERA did not change the HECM loan limits in parts of Hawaii that had loan limits higher than $417,000 prior to HERA. As a result, borrowers in those areas whose house values are greater than the pre-HERA loan limits will see no change in their up-front mortgage insurance premiums due to HERA.


$200,000 would see no change in the up-front costs because they would pay the same 2 percent of the maximum claim amount (the same as before HERA). A borrower whose house is valued at greater than $200,000 would expect a decrease in up-front costs due to the decreased origination fee for amounts greater than $200,000 and the fee cap of $6,000. For an example, see borrower D, whose house value is $300,000, in table 1.

• Maximum claim amount increases: For borrowers whose maximum claim amount increases because their house values are greater than the prior local HECM loan limit, the change to up-front costs is more complex. All borrowers in this category will pay more in up-front mortgage insurance premiums because premiums are calculated based on the entire maximum claim amount. However, some borrowers may pay more in origination fees, while others will pay less. When combining these two costs, the total up-front costs could increase, decrease or remain the same. For example, borrowers A, B, and C in table 1 each own houses valued at $300,000 that are located in counties in which prior HECM loan limits varied from $200,000 to $290,000. Each borrower would see different effects in up-front costs.

See appendix III for a more complete explanation of how up-front costs will change for borrowers with different characteristics.

Table 1: Change in Up-front Costs for $300,000 Houses in Various Locations

Borrower

Prior local HECM loan

limit Maximum claim

amount

Up-front mortgage insurance premium

Origination fee

Total up-front costs

Change in up-front costs

A $200,000 Before HERA $200,000 $4,000 $4,000 $8,000 $3,000 more

After HERA 300,000 6,000 5,000 11,000

B 275,000 Before HERA 275,000 5,500 5,500 11,000 No change

After HERA 300,000 6,000 5,000 11,000

C 290,000 Before HERA 290,000 5,800 5,800 11,600 $600 less

After HERA 300,000 6,000 5,000 11,000

D 325,000 Before HERA 300,000 6,000 6,000 12,000 $1,000 less

After HERA 300,000 6,000 5,000 11,000

Source: GAO.


To illustrate the potential effect of the HERA provisions on borrowers, we compared the actual maximum claim amounts, up-front costs (origination fee plus the up-front insurance premium), and loan funds available for HECM borrowers in 2007 to what their maximum claim amounts, up-front costs, and loan funds available would have been had the HERA provisions been in place.17 Overall, we found that nearly 27 percent of borrowers would have paid more in up-front costs, 46 percent would have paid less, and 27 percent would have paid the same (see fig. 8). The amount and direction of the changes to up-front costs and loan funds available primarily depended on house value and whether a borrower would have benefited from an increase in loan limit (about 28 percent of 2007 HECM borrowers’ homes were valued at more than the prior loan limit and would have seen their maximum claim amounts increase because of HERA’s increase in the loan limit).

Most 2007 HECM Borrowers Would Have Paid the Same or Less in Up-front Costs under the HERA Provisions, and Most Borrowers Would Have Had the Same or More Loan Funds Available

Our analysis of up-front costs broken down by its two components is as follows:

• Origination fees: About 24 percent of 2007 borrowers would have paid more in origination fees, 49 percent would have paid less, and 27 percent would have paid the same amount. Increases in origination fees were due either to the $500 increase in the minimum origination fee (about 17 percent of all borrowers) or to the increased loan limits (about 6 percent of all borrowers). Borrowers who would have paid less in origination fees had maximum claim amounts greater than $200,000, which means they would have benefited from the decrease in the origination fee for the portion of the maximum claim amount greater than $200,000, the $6,000 origination fee cap, or both.

• Up-front mortgage insurance premium: Twenty-eight percent of 2007 HECM borrowers would have paid more in up-front mortgage insurance premiums due to increases in the loan limit, while 72 percent of borrowers would have paid the same amount, generally because the size of their loans was limited by the value of their homes and not the HECM loan limit.18

17Loan funds available for a HECM, is the loan amount after applying the principal limit factor to the maximum claim amount and deducting the up-front costs and servicing fee set-aside (see fig. 5).

18For some Hawaii borrowers, the up-front mortgage insurance premium would have been the same because their loans were capped by loan limits that did not change due to HERA.


Changes in the loan limits and up-front fees would have affected the loan funds available to most 2007 borrowers. Borrowers whose maximum claim amount would have increased because of an increase in loan limit would have paid a higher up-front mortgage insurance premium, regardless of how much of their available loan funds they chose to access. Because this analysis assumed that HECM borrowers financed the up-front costs in the loan, any increase or decrease in the up-front costs affects the amount of loan funds that are available to them. Our analysis—which assumes that borrowers financed their up-front costs—shows that had the HERA provisions been in place at origination for 2007 HECMs, approximately 56 percent of borrowers would have had more loan funds available, 17 percent would have had less loan funds available, and 27 percent would have had the same amount available (see fig. 8).

Specifically,

• 28 percent of borrowers would have had more loan funds available, primarily due to the increase in loan limit;

• about 28 percent of borrowers would have had more loan funds available due solely to a decrease in their up-front fees;

• 17 percent of borrowers would have had a smaller amount of loan funds available due solely to an increase in their up-front fees; and

• 27 percent of borrowers would have experienced no change in the amount of loan funds available because their up-front fees and loan limits remained the same.


Figure 8: Number of 2007 HECM Borrowers Affected by HERA Provisions

aFor 18,847 of these borrowers, the maximum claim amount increased. For the other 28,015 borrowers, the maximum claim amount remained the same.

Additionally, figure 8 shows the number of 2007 borrowers within the various categories and figure 9 shows the average changes in up-front costs and loan funds available for each category of borrower. Borrowers with the largest increases in their maximum claim amounts on average would have the largest percent increases in up-front costs (see fig. 9). Borrowers with no increase in their maximum claim amount, who have a change to up-front costs, will have a corresponding change in loan funds available that are equal in size but opposite in direction. For example a borrower with a $200 decrease in up-front costs will have a $200 increase in loan funds available and a borrower with a $300 increase in up-front costs will have a $300 decrease in loan funds available.

Source: GAO.

Change in loan funds available

Change in up-front costs

Total

27,064(27%)

27,554(27%)

46,862(46%)

17,303(17%)

0

0

0

27,548(27%)

0

9,761(10%)

6(less than

1%)

46,862(46%)

a

Total 101,480(100%)

17,303(17%)

27,548(27%)

56,629(56%)

Decrease

Decrease

Increase

Increase

No change

No change


Figure 9: Average Changes in Maximum Claim Amounts, Up-front Costs, and Loan Funds Available for 2007 HECM Borrowersa

aThis analysis did not include refinanced HECMs, but a separate analysis of refinanced HECMs showed similar changes in maximum claim amounts, up-front costs, and loan funds available within the different categories of borrowers. bFor these borrowers, the maximum claim amount increased. cFor these borrowers, the maximum claim amount remained the same.

Source: GAO.

Number of2007 HECMs

Changein up-frontcosts

Change inloan fundsavailable

Averageoriginalmaximum claim amount

Averageincrease inmaximumclaim amount

Averagedifference in up-front costs

Averagedifference in loan fundsavailable

9,761 $240,008

90,760

260,875

162,592

350,401

270,918

$64,368

0

20,292

0

39,477

0

$1,508(11%)

$42,560(28%)

17,303402

(5%)-402

(-1%)

60

(0%)14,232

(8%)

27,5480

(0%)0

(0%)

18,847 b

c

-422(-2%)

27,883(12%)

28,015-733

(-4%)

733(less than

1%)

Increase

Decrease

No change

Borrowers May Be Affected by Other Factors, Such as Lender Margin Rates and Counseling Fees

Increased lender margin rates stemming from HERA’s change to the origination fee calculation could reduce loan funds available to borrowers. At loan origination, the expected interest rate HUD uses to determine the portion of the maximum claim amount that will be made available to the borrower includes the 10-year Treasury rate plus the fixed lender margin rate. Our survey of HECM lenders indicates that some lenders have raised their margin rates modestly to compensate for HERA’s limitations on the origination fee; however, we did not receive a sufficient number of responses to reliably estimate the median increase in margin rate for the


population.19 To illustrate the impact of a modest increase in margin-rate on borrowers, we applied a 0.25 percentage point increase to borrowers who took out HECMs in 2007. We found that these borrowers would have seen a 3 percent average decrease in loan funds available as a result of the higher margin rate.20 A comparison of HUD data on HECMs originated within the first 3 months of HERA’s implementation with data from the same 3 months from the prior year indicates that average margin rates were higher after HERA but that the overall average HECM expected interest rates were essentially the same. This outcome resulted from declines in 10-year Treasury rates offsetting increases in lender margin rates.

In addition, more borrowers, as well as prospective borrowers who ultimately do not obtain a HECM, may need to pay counseling fees. Provisions in HERA prohibit lenders from paying for this counseling but allow HUD to use a portion of HECM mortgage insurance premiums for this purpose. HUD officials said that they have not exercised this authority because the resulting reduction in premium income would affect the subsidy rate of the program adversely and potentially require appropriations. Because HUD did not implement this provision, more borrowers and prospective borrowers may need to pay counseling fees themselves. For borrowers who do eventually obtain a HECM, the fee can be financed in the loan. Prospective borrowers who do not qualify for a HECM or who choose not to proceed with the loan after counseling may have to pay for counseling out of pocket. HUD’s recent announcement that it will provide approximately $8 million in grant funds for HECM counseling in 2009 may mitigate any negative impact the HERA changes may have on seniors’ ability to obtain HECM counseling.

19Of the lenders that responded to the survey question, we estimate that 48 percent increased margin rates for HECMs offered by their institution by at least 0.25 percentage points. This estimate has a margin of error of within plus or minus 13 percentage points.

20All other things being equal, an increase in margin rate causes a decrease in loan funds available (see fig. 5). This analysis uses the 10-year Treasury rate that was available to the borrower at loan origination.


HUD has taken or planned steps to enhance its analysis of the HECM program’s financial performance. However, HUD’s recent estimates of program costs indicate weaker performance than previously estimated, primarily due to more pessimistic assumptions about long-term house price trends. Additionally, higher loan limits enacted under HERA and the American Recovery and Reinvestment Act of 2009 (ARRA) could increase HUD’s financial risk.

HUD Has Enhanced Its Analysis of HECM Program Costs but Changes in House Price Trends and Higher Loan Limits Have Increased HUD’s Risk of Losses

HUD Is Taking Steps to Improve its Analysis of the HECM Program’s Financial Performance

To estimate the cost of the HECM program, HUD uses a model to project the cash inflows (such as insurance premiums paid by borrowers) and cash outflows (such as claim payments to lenders) for all loans over their expected duration. HUD’s model is a computer-based spreadsheet that incorporates assumptions based on historical and projected data to estimate the amount and timing of insurance claims, subsequent recoveries from these claims, and premiums and fees paid by borrowers. These assumptions include estimates of house price appreciation, interest rates, average loan size, and the growth of unpaid loan balances. HUD inputs its estimated cash flows into OMB’s credit subsidy calculator, which calculates the present value of the cash flows and produces the official credit subsidy rate for a particular loan cohort. A positive credit subsidy rate means that the present value of the cohort’s expected cash outflows is greater than the inflows, and a negative credit subsidy rate means that the present value of the cohort’s expected cash inflows is greater than the outflows. To budget for a positive subsidy an agency must receive an appropriation. HUD also uses the cash flow model to annually estimate the liability for loan guarantees (LLG), which represents the net present value of future cash flows for active loans, taking into account the prior performance of those loans. HUD estimates the LLG for individual cohorts as well as for all cohorts combined.21 The LLG is a useful statistic because unusual fluctuations in the LLG can alert managers to financial risks that require further attention.

21The LLG does not include cash flows that have already occurred.


HUD in recent years has enhanced its cash flow model for the HECM program. In 2007, the HUD Office of Inspector General’s (OIG) annual audit of FHA’s financial statements cited a material weakness in the cash flow model FHA used to generate credit subsidy estimates for the HECM program.22 Among other things, the audit noted technical errors in the model, significant discrepancies between projected and actual cash flows, and a lack of supporting documentation for certain modeling decisions. Partly in response to the OIG audit, HUD made a number of improvements to both the model and its supporting documentation, and in 2008 the HUD OIG eliminated the material weakness.23 For example, HUD improved the methodology it uses for its cash flow model. In the past, HUD used historical averages for termination and recovery rates for projecting cash flows. In 2008, HUD began to incorporate forecasts of national house price appreciation and interest rates from IHS Global Insight, an independent source for economic and financial forecasts, into its modeling.24 Additionally, HUD improved the way it estimates the growth of unpaid principal balances, which HUD uses to calculate the LLG. In the past, HUD used both active and terminated loans to generate this estimate. Since 2008, HUD has included only active loans to generate this estimate, which is more appropriate because the LLG represents the expected future cash flows of currently active loans. HUD also developed a master database of loan-level information to support the HECM cash flow model. Previously, HUD staff had to draw on data from multiple sources, which increased the chance of analytical errors. Finally, HUD made a number of enhancements to its documentation of estimation processes, including how macroeconomic projections are incorporated into the cash flow model.

HUD plans to subject the HECM program to an annual actuarial review, which should provide additional insight into the program’s financial condition. Such a review would likely assess if program reserves and funding were sufficient to cover estimated future losses, as well as the sensitivity of this analysis to different economic and policy assumptions. Historically, the HECM program has not had a routine actuarial review because it was supported by the General Insurance and Special Risk

22HUD OIG, Audit of the Federal Housing Administration’s Financial Statements for

Fiscal Years 2007 and 2006, 2008-FO-0002 (Washington, D.C., Nov. 8, 2007).

23HUD OIG, Audit of the Federal Housing Administration’s Financial Statements for

Fiscal Years 2008 and 2007, 2009-FO-0002 (Washington, D.C., Nov. 7, 2008).

24IHS Global Insight is a private company that forecasts a wide range of financial and economic indicators.


Insurance Fund (GI/SRI) Fund, which does not have such a review requirement.25 However, as of fiscal year 2009, the HECM program is in the Mutual Mortgage Insurance (MMI) Fund, which is statutorily required to receive an independent actuarial review each year and includes FHA’s largest mortgage insurance program.26 HUD officials told us that future actuarial reviews of the MMI Fund will include a separate assessment of the HECM program.

HUD also is considering producing credit subsidy re-estimates for the HECM program.27 As discussed later in this report, HUD has generated credit subsidy estimates for individual HECM cohorts for several years. However, HUD officials told us that, until recently, they did not have the data necessary to produce subsidy re-estimates for HECMs. Specifically, the officials noted that for HECM cohorts prior to 2009, assets for HECMs were aggregated with assets from other programs in the GI/SRI Fund and not accounted for separately. HUD officials said that they are now accounting for HECM assets separately, which will enable them to produce re-estimates for the HECM program. Re-estimates can highlight cohorts that are not expected to meet original budget estimates. This information could help inform future actions to manage HUD’s insurance risk and control program costs.

Prior Cost Estimates Indicated That the HECM Program Was Profitable but Current Estimates Forecast Losses, Primarily Due to Revised House Price Assumptions

HUD’s most recent estimates of two important financial indicators for the HECM program—the credit subsidy rate and the LLG—suggest weaker financial performance than previously estimated, largely due to more pessimistic house price assumptions. All other things being equal, lower house price appreciation can increase HUD’s insurance losses because it makes it less likely that the value of the home will cover the loan balance. Analyses by HUD have found that the financial performance of the HECM program is sensitive to long-term trends in house prices. HUD officials told us that HECM program performance is less sensitive to short-term price

25Pursuant to congressional directives, HUD submitted reports on the HECM program in 1995, 2000, and 2003 (P.L. 100-242 and P.L. 106-569). These reports included actuarial reviews of the program.

2612 U.S.C. Sec. 1708(a)(4). HECMs originated in fiscal year 2009 and beyond will be accounted for under the MMI Fund. Loans originated in or before fiscal year 2008 will be accounted for under the GI/SRI Fund.

27Currently, HUD includes HECMs in its credit subsidy re-estimates for the GI/SRI Fund as a whole.


declines because borrowers with HECMs, unlike those with traditional forward mortgages, do not have an incentive to terminate (or default on) their loans when prices fall.

HUD has made credit subsidy estimates for HECM cohorts from 2006 forward. Because the HECM program was relatively small prior to 2006, HUD did not produce separate subsidy estimates for the HECM program but included HECMs in its estimates of subsidy costs for the GI/SRI Fund as a whole. For the 2006 through 2009 HECM cohorts, HUD estimated negative subsidy rates ranging from - 2.82 percent in 2007 to -1.37 percent in 2009 (see fig. 10). However, for the 2010 cohort, HUD estimated a positive subsidy rate of 2.66 percent. Because HUD is expecting to insure about $30 billion in HECMs in 2010, this rate corresponds to a subsidy cost of $798 million. As required by the Federal Credit Reform Act, the President’s budget for fiscal year 2010 includes a request for this amount.28

28On July 17, 2009, the House Committee on Appropriations approved a fiscal year 2010 appropriations bill for HUD. The bill contained a provision directing HUD to adjust, as necessary, the principal limit factor for new loans to ensure that the program operates without a subsidy. The provision also prohibited HUD from reducing a borrower’s principal limit factor below 60 percent. If enacted, this provision would eliminate the need for the $798 million budget request but would also reduce the loan funds available to some borrowers.


Figure 10: HECM Credit Subsidy Rates, Fiscal Years 2006 through 2010

-3

-2

-1

0

1

2

3

20102009200820072006

Subsidy rate (percentage)

Source: GAO analysis of OMB federal credit supplement data (2006 through 2010).

Cohort

Yes

No

Subsidy required?

-1.74%

-2.82%

-1.90%

-1.37%

2.66%

HUD officials told us that the positive subsidy rate for fiscal year 2010 largely was due to incorporating more conservative assumptions about long-term house price trends than had been used for prior cohorts. For budgeting purposes, the Administration decided to use more modest appreciation rates than the private sector forecasts HUD typically uses. Specifically, the house price appreciation rates used were 0.5 percent greater than the forecasted inflation rates. HUD officials told us that if they had used IHS Global Insight projections to develop the fiscal year 2010 credit subsidy estimate, there would be no need for an appropriation because the credit subsidy rate would be negative.

HUD also has estimated the LLG for the HECM program since 2006. As shown in figure 11, HUD’s original LLG estimates grew substantially from 2007 to 2008, increasing from $326 million to $1.52 billion. According to FHA’s financial statements for fiscal years 2007 and 2008, the increase was primarily due to the lower house price appreciation projections used in the 2008 analysis.29 The report noted that lower appreciation rates result in

29HUD, FHA Annual Management Report Fiscal Year 2007 (Washington, D.C., 2007) and FHA Annual Management Report Fiscal Year 2008 (Washington, D.C., November 2008).


lower recoveries on mortgages assigned to HUD, which in turn increases HUD’s liability.

Figure 11: Loan Liability Guarantee for the HECM Program, Fiscal Years 2006 through 2008

0

200

400

600

800

1,000

1,200

1,400

1,600

200820072006

Dollars (in millions)

Source: GAO analysis of FHA financial statements (2006 through 2008).

Fiscal year

In September 2008, HUD analyzed the sensitivity of the 2008 LLG estimate for the HECM program as a whole to different assumptions, including alternative house price scenarios. HUD examined the impact of house price appreciation that was 10 percent higher and 10 percent lower than the baseline assumptions from IHS Global Insight for fiscal years 2009 through 2013. (For example, for a baseline assumption of 4 percent house price appreciation, the lower and higher scenarios would have been 3.6 percent and 4.4 percent, respectively.) HUD estimated that the more pessimistic assumption increased the LLG from $1.52 billion to $1.78 billion, while the more optimistic assumption reduced the LLG to $1.27 billion.


When estimating future costs for all HECMS, HUD assumes that the property value at loan origination is equal to the maximum claim amount. For loans in which the property value is more than the HECM loan limit, this approach results in a conservative assumption about the amount of home equity available at the end of the loan to cover the loan balance. In these cases, the actual home value at the end of the loan is likely to be more than what HUD assumes and therefore more likely to exceed the loan balance at the end of the loan. According to HUD, because of this conservative approach to estimating costs, the HECM program does not rely on loans with property values that exceed the maximum claim amount to operate on a break-even basis over the long-run.

HUD Uses a Conservative Approach in Estimating Program Costs, but Higher Loan Limits May Increase the Potential for Losses

Higher loan limits enacted under HERA and ARRA may make HUD’s approach less conservative by reducing the proportion of loans for which the property value exceeds the maximum claim amount. This scenario is especially likely in locations that previously had relatively low local loan limits (reflecting their lower home values) but are now subject to the higher national limit. To illustrate, consider a 65-year-old HECM borrower with a $400,000 home whose loan limit prior to HERA was $250,000 (see fig. 12). In this scenario, the maximum claim amount would be the same as the loan limit because the maximum claim amount is defined as the lesser of the loan limit or the home value. However, if the loan limit for the same borrower is increased to the HERA-authorized level of $417,000, the maximum claim amount is the same as the home value ($400,000). As figure 12 shows, when a borrower’s maximum claim amount is capped by the loan limit, the maximum claim amount can be substantially lower than the value of the home. All other things being equal, the potential for losses is low in this scenario because the projected loan balance is likely to remain less than the projected home value after the lender assigns the loan to HUD.30 In contrast, when the maximum claim amount is capped by the home’s value, the difference between the projected loan balance and the projected home value is smaller. The potential for losses is higher with such a loan because the projected loan balance is more likely to exceed the projected home value. As also shown in figure 12, when this effect is combined with declining home prices, the potential for losses increases.

30As previously noted, lenders may assign the loan to HUD when the loan balance reaches 98 percent of the maximum claim amount. At that point, HUD takes over servicing of the loan and is responsible for covering any losses.


Figure 12: Illustration of How Increasing the HECM Loan Limit Could Increase HUD’s Losses

0

200

400

600

800

1,000

1,200

1,400

30282624222018161412108642



$250,000

max claim amountLoan

assignedto HUD

Property value

Loan balance

0

200

400

600

800

1,000

1,200

1,400

30282624222018161412108642



$400,000

max claim amountLoan

assignedto HUD

Property value

Loan

bala

nce

Pre-HERA Post-HERA

No loss area (property value exceeds loan balance)

Loss area (loan balance exceeds property value)

Short-term decline in property values at time of origination

Source: GAO.

Note: The trend line for property value assumes house price appreciation of 4 percent annually—this is the rate HUD assumes in its pricing of the HECM loan product. The initial loan amount is based on a $400,000 home, as well as an expected interest rate of 5.43 percent and a borrower age of 65, which corresponds to a principal limit factor of 64.9 percent. The trend line for loan balance assumes that the borrower immediately drew down 100 percent of the available loan funds.

Studies by HUD and others have noted that HECM loans for which the home value exceeds the maximum claim amount have a positive impact on the program’s financial performance but also have noted the potential negative impact of raising the loan limit. When the HECM program started in 1990, HUD developed a statistical model to estimate borrower payments and insurance risk. HUD’s technical explanation of the model acknowledges that future expected losses are smaller for HECMs with a


maximum claim amount capped by the loan limit, as compared with HECMs with a maximum claim amount equal to the home value.31 Similarly, actuarial reviews of the HECM program—conducted in 1995, 2000, and 2003—concluded that the negative net liability of the HECM program resulted from homes valued at more than the HECM loan limit cross-subsidizing those valued at less than the limit.32 The 2003 actuarial review also examined how the financial condition of the HECM program would have been affected had a higher, national loan limit been in place when existing HECMs were originated. The analysis found that the higher loan limits would have reduced the expected net liability of the HECM program from -$54.0 million to -$11.4 million. This finding is consistent with a Congressional Budget Office (CBO) analysis of a 2007 legislative proposal to increase the HECM loan limit to $417,000 nationwide. CBO concluded that the increase would reduce HUD’s credit subsidy rate for the 2008 cohort of loans from -1.9 percent to -1.35 percent.33

The percentage of HECMs with maximum claim amounts capped by the loan limit has declined in recent years (see fig. 13). Since the inception of the program, this percentage has ranged from 24 percent to 47 percent. However, this proportion has declined in recent years, dropping from 42 percent in fiscal year 2006 to 25 percent in fiscal year 2008. Furthermore, HUD data show that this proportion dropped to 18 percent for the first 4 months of fiscal year 2009, likely due in part to the higher loan limit.

31HUD, The FHA Home Equity Conversion Mortgage Insurance Demonstration: A Model

to Calculate Borrower Payments and Insurance Risk (Washington, D.C., October 1990).

32HUD, Evaluation of the Home Equity Conversion Mortgage Insurance Demonstration:

Report to Congress. (Washington, D.C., 1995); Evaluation Report of FHA’s Home Equity

Conversion Mortgage Insurance Demonstration. Prepared by Abt. Assoc. (Washington, D.C., Mar. 31, 2000); Refinancing Premium, National Loan Limit, and Long-Term Care

Premium Waiver For FHA’s HECM Program (Final Report). Prepared by Abt. Assoc. (Washington, D.C., May 2003).

33CBO, Cost Estimate: FHA Modernization Act of 2007 (Washington, D.C., Oct. 12, 2007).


Figure 13: Percentage of HECMs with Maximum Claim Amounts Limited by the Program Limit

Percentage

Fiscal year

Maximum claim amounts capped by HECM loan limit

Maximum claim amounts capped by value of property

Source: GAO analysis of HUD data.

0

10

20

30

40

50

60

70

80

90

100

20092008200720062005200420032002200120001999199819971996199519941993199219911990

2531424136343131

2425272932393839

4748

40

18

7569

58596466

6969767573

7168616261

5352

82

60

Note: Fiscal year 2009 data include HECMs insured from October 1, 2008, through January 31, 2009.

HUD officials acknowledged that a reduction in the proportion of loans with maximum claim amounts capped by the loan limit could have a negative effect on the program’s financial performance. However, they also indicated that their conservative approach to estimating program costs mitigates the associated risks.


We provided a draft of this report to HUD for its review and comment. In comments provided to us in an e-mail, HUD concurred with our report and provided a technical comment, which we incorporated into the report.

Agency Comments and Our Evaluation

We are sending copies of this report to interested congressional parties,

the Secretary of the Department of Housing and Urban Development, and other interested parties. In addition, the report will be available at no charge on our Web site at http://www.gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. If you or your staff has any questions about this report, please contact me at (202) 512-8678 or [email protected]. GAO

Mathew J. Scirè

contact information and staff acknowledgments are listed in appendix IV.

Director, Financial Markets y Investment and Communit



Appendix I: Objectives, Scope, and

Methodology

Page 38 GAO-09-836

Appendix I: Objectives, Scope, and Methodology

Our objectives were to examine (1) how the Housing and Economic Recovery Act of 2008 (HERA) changes to the Home Equity Conversion Mortgage (HECM) program and other factors have affected HECM lenders’ planned participation in the reverse mortgage market, (2) the extent to which HERA’s changes to HECM origination fees and loan limits will affect costs to borrowers and the loan amounts available to them, and (3) Department of Housing and Urban Development’s (HUD) actions to evaluate the financial performance of the HECM program, including the potential impact of loan limit and house price changes.

To address these objectives, we reviewed laws, regulations and guidance relevant to the HECM program, including provisions in HERA, the American Recovery and Reinvestment Act of 2009 (ARRA), and HUD handbooks and mortgagee letters. We also spoke with agency, industry, and nonprofit officials, including those at HUD, Ginnie Mae, Fannie Mae, the National Reverse Mortgage Lenders Association (NRMLA), the Mortgage Bankers Association (MBA), and AARP.

To determine how HERA’s provisions have affected lenders’ planned participation in the reverse mortgage market, we spoke with industry and nonprofit officials—including those at Ginnie Mae, Fannie Mae, AARP, NRMLA, and MBA—to understand how recent legislative and economic changes were affecting the industry. To more specifically identify the influence of legislation and economic factors on HECM lenders, we conducted a Web-based survey of a random probability sample of the 2,779 lenders that originated HECMs on a retail basis in fiscal year 2008. We used HUD records of HECM-certified lenders making at least one such loan in fiscal year 2008, and supplemented HUD’s loan company officer contact information with names and e-mail addresses of officers at those lenders in our sample who also had memberships in NRMLA. For the remaining sampled lenders for which we lacked contact information, we made telephone calls to identify the most appropriate recipient for our survey invitation.

We drew a stratified sample, allocating our selections across three groups defined by the number of HECMs made in fiscal year 2008, sampling from the groups with larger lenders at a higher rate than from the groups with smaller lenders (see table 2). We sampled all 51 members of the stratum with the largest lenders (300 or more loans). We sampled so few (30) and received so few usable responses (8) from the stratum with the smallest lenders (1 to 9 loans), that we considered this a nongeneralizable sample and excluded it from our quantitative analysis. In addition, lenders in the smallest lender stratum account for less than 5 percent of all loans, and

Reverse Mortgages


Methodology

thus would not influence overall estimates very much. Responses from the smallest lenders stratum were used only as case study examples in our analysis.

Table 2: Survey Population and Sample Dispositions

Stratum Population

Fiscal year 2008

HECMs Originalsample

Total responses

Ineligibleresponses

Usable responses

Non- responses ea

Response rateb

Very small (1-9 HECM loans in 2008)c 1,837 5,549 30 10 2 8 20 .80 33.3

Small (10 to 59 loans in 2008)

669 16,132 170 83 3 80 87 .96 48.8

Medium (60 to 299 loans in 2008)

222 27,247 113 66 13 53 47 .80 58.4

Large (300 or more loans in 2008)

51 63,087 51 41 2 39 10 .95 80.8

Totalc 2,779 112,015 364 200 20 180 164 .91 56.9%c

Source: GAO. ae = estimated eligibility rate of nonresponding sample cases whose eligibility is unknown. bResponse rate is unweighted, as defined by AAPOR RR3. c“Very small” stratum not included in overall response rate calculations.

To help develop our questionnaire, we consulted with an expert at NRMLA. We pretested our draft questionnaire to officials at three HECM lenders in our population and made revisions to it before finalization. Legal and survey research specialists in GAO also reviewed the questionnaire.

Before the survey, in early March 2009, NRMLA sent letters to those lenders in our sample who were also members in that organization, endorsing our survey and encouraging response. In March 2009, we sent e-mails with links to our Web questionnaire and unique login information to each member of our sample with valid e-mail addresses. For sampled companies for which we were unable to obtain working e-mail addresses, we mailed paper versions of the questionnaires. Nonresponding lenders were sent additional e-mails or copies of questionnaires from March through May. We also made telephone calls in April to nonrespondents encouraging them to respond. Our survey closed in early May 2009. We



Methodology

received a total of 180 usable responses, for an overall response rate of 57 percent.1 The “weighted” response rate for the survey, which takes into account the relative numbers of lenders in the population that sampled lenders in each of our three size strata had to represent, was 53 percent. The most common reason for ineligibility among our sample firms was closure, merger, or other discontinuation of business in the reverse mortgage industry.

Because we followed a probability procedure based on random selections, our sample is only one of a large number of samples that we might have drawn. Since each sample could have provided different estimates, we express our confidence in the precision of our particular sample’s results as a 95 percent confidence interval (e.g., plus or minus 10 percentage points). This is the interval that would contain the actual population value for 95 percent of the samples we could have drawn. As a result, we are 95 percent confident that each of the confidence intervals in this report will include the true values in the study population. Unless otherwise noted, our estimates have margins of error of plus or minus 10 percentage points or less at the 95 percent confidence interval.

In addition to sampling error, the practical difficulties of conducting any survey may introduce other errors:

1. Nonresponse—bias from failing to get reports from lenders whose answers would have differed significantly from those who did participate.

2. Coverage—failure to include all eligible HECM lenders in the list from which we sampled, or including ineligible firms.

3. Measurement—errors in response.

4. Data processing. We took steps in developing the questionnaire, collecting the data, and analyzing them to minimize such errors. For example, our pretesting and expert reviews of the questionnaire resulted in question changes that reduced the possibility of measurement error, and all data processing and analysis programming was verified by independent analysts. In addition,

1We used the response rate definition “RR3,” as defined by the American Association for Public Opinion Research in “Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys,” http://www.aapor.org/uploads/Standard_Definitions_07_08_Final.pdf, p. 35.


http://www.aapor.org/uploads/Standard_Definitions_07_08_Final.pdf


Methodology

we followed up on some unlikely answers by recontacting sampled lenders or conducting followup research on them to edit erroneous answers and declare some firms ineligible for our survey, thereby reducing measurement and coverage error.

To assess the risk of nonresponse bias, we compared the response rates of lenders across categories of two characteristics that might be related to our key variables—the effect of HERA changes and other factors on the likelihood of continuation of HECM lending in the future. The two characteristics known for both respondents and nonrespondents were the number of years the lender had been offering HECMs and the state in which the lender’s home office is located, from which we could develop a measure of size of loan activity in each state by summing the number of loans made by lenders whose home offices were in a given state. We found no statistically significant association between these two characteristics and the likelihood of response.2 Although this does not eliminate the possibility of nonresponse bias, we found no evidence of bias based on our analysis of this available data.

To determine the effect of the HERA provisions on HECM borrowers, we examined changes in the up-front mortgage insurance premium, origination fee, and loan funds available to borrowers. The up-front mortgage insurance premium is 2 percent of the maximum claim amount. HERA did not change this rate, but because of HERA’s change to the HECM loan limit, some borrowers may be eligible for larger loans and therefore have higher maximum claim amounts. Since the premium is calculated based on the maximum claim amount, these borrowers will pay a higher up-front mortgage insurance premium than they would have prior to HERA. Before HERA, the origination fee was calculated as 2 percent of the maximum claim amount with a minimum fee of $2,000. HERA changed the calculation of the origination fee to 2 percent of the first $200,000 of the maximum claim amount plus 1 percent of the maximum claim amount over $200,000, with a maximum fee of $6,000. In implementing HERA, HUD also increased the minimum origination fee by $500 to $2,500.

2We grouped the sample lenders into two groups defined by when they were first approved to offer HECM loans (before 2000 and 2000 or later), and found no association between the year category and whether the lender responded. In addition, we compared sampled lenders in the 5 states with the most HECM loans to the rest of the sampled lenders in the remaining states and also found no association with whether the lender responded.



Methodology

We used two different approaches to assess the impact of the HERA changes. First, we performed a mathematical analysis showing the difference between the up-front costs before and after HERA. Specifically, we derived equations for calculating pre-HERA and post-HERA up-front costs for borrowers with maximum claim amounts in different ranges ($0 to $100,000; $100,000 to $125,000; $125,000 to $200,000; $200,000 to $400,000; and $400,000 to $625,500).3 For each range, we subtracted the pre-HERA equation from the post-HERA equation to derive an equation for calculating the change in up-front costs due to the HERA provisions. We then used these equations to calculate the potential change in up-front costs in dollars terms. We did this analysis separately for cases in which the maximum claim amount would increase under HERA and cases in which the maximum claim amount would remain the same. Appendix III shows the details of this analysis.

Second, we applied the HERA changes to HUD loan-level data for HECMs that borrowers obtained in calendar year 2007. We compared the results to the actual up-front costs and loan funds available for these borrowers. To perform this analysis, we obtained data from HUD’s Single-family Data Warehouse. We assessed the reliability of these data by (1) reviewing existing information about the data and the system that produced them, (2) interviewing HUD officials knowledgeable about the data, and (3) performing electronic testing of required data elements. We determined that the data we used were sufficiently reliable for the purposes of this report. As shown in table 3, the universe of 2007 HECMs used in our analysis included 101,480 loans.4 We applied the $417,000 national loan limit and HERA’s changes to the origination fee calculation to the 2007 HECMs. For each borrower, we calculated the new maximum claim amount, origination fee, up-front mortgages insurance premium, and loan funds available under the HERA rules and compared our results to the actual 2007 values. We summarized our results by calculating the average changes in these amounts.

3We used $625,500 as a maximum because ARRA resulted in an increase in the HECM loan limit to that amount.

4The 101,480 loans did not include 6,664 HECMs obtained in 2007 to refinance an existing HECM. We analyzed refinanced HECMs separately because the calculation for the mortgage insurance premium is different for these loans than for new HECMs. Our results for the analysis of refinanced HECMs were similar to those for new HECMs.



Methodology

Table 3: Universe of 2007 HECMs by Home Value and Maximum Claim Amount Category

Home value Maximum claim amount

equals loan limitMaximum claim amount

equals home value Total

Under $125,000 0 17,303 17,303(17%)

$125,000 to $200,000 6 27,548 27,554)(27%)

$200,001 to $399,999 14,535 27,474 42,009(41%)

$400,000 or more 14,433 181 14,614(14%)

Total 28,974(29%)

72,506(71%)

101,480(100%)

Source: GAO analysis of HUD data.

Note: Percentages do not add to 100 because of rounding.

To illustrate the potential effect of modest margin rates increases stemming from HERA’s change to the origination fee calculation, we applied a 0.25 percentage point increase to the margin rate for the 2007 HECMs adjusted to reflect the HERA provisions.5 We determined the resulting changes in the loan funds available to borrowers using HUD’s table of principal limit factors. To provide perspective on the HERA-related margin rate changes, we compared margin rates from a 3 month period 1 year prior to the implementation of HERA (November 2007 through January 2008) to the margin rates from the 3 month period after the implementation of HERA (November 2008 through January 2009).

To examine HUD’s actions to evaluate the financial performance of the HECM program, we reviewed HUD’s budget estimates for the HECM program for fiscal years 2005 through 2010. We also compiled and analyzed financial performance information about the HECM program, including the liability for loan guarantee (LLG) and credit subsidy

5As discussed in the body of this report, our survey of HECM lenders indicated that some lenders have raised their margin rates modestly to compensate for HERA’s limitations on the origination fee. We did not receive a sufficient number of responses to reliably estimate the average increase in margin rate for the population. Of the lenders that responded to the survey question, we estimate that 48 percent increased margin rates for HECMs offered by their institution by at least 0.25 percentage points. This estimate has a margin of error of within plus or minus 13 percentage points.



Methodology

estimates. For example, we examined the Federal Housing Administration’s (FHA) Annual Management Reports (2005, 2006, 2007, and 2008), which include FHA’s annual financial statements; HUD Office of the Inspector General (OIG) audits of FHA’s financial statements (2005, 2006, 2007, and 2008); actuarial reviews of the HECM program (1995, 2000, and 2003); and Congressional Budget Office cost estimates relevant to the HECM program. We also reviewed other analyses HUD has conducted of program costs, such as the sensitivity of estimated cash flows to alternative economic assumptions. We interviewed FHA officials about their budget estimates and program analyses. Additionally, we reviewed information about HUD’s HECM cash flow model, including a technical explanation of the model published in 1990 and recent changes to the model. We also reviewed historical house price appreciation rates from the Federal Housing Finance Agency and projected house price appreciation rates from IHS Global Insight. To examine the percentage of HECMs with maximum claim amounts capped by the loan limit, we analyzed loan-level data on HECMs from HUD’s Single-family Data Warehouse. As noted earlier, we determined that the data we used were sufficiently reliable for this analysis. In addition, we reviewed federal agency standards for managing credit programs, such as those contained in the Federal Credit Reform Act (FCRA), related Office of Management and Budget requirements and instructions, and Federal Accounting Standards Advisory Board guidance. Finally, we interviewed HUD OIG officials, industry participants, and mortgage market analysts.

We conducted this performance audit from September 2008 through July 2009, in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.


Appendix II: Impact of

the American R

of 20

Loan Limit Increase in

ecovery and Reinvestment Act

09 on HECM Lenders

Page 45 GAO-09-836

Appendix II: Impact of Loan Limit Increase in the American Recovery and Reinvestment Act of 2009 on HECM Lenders

The American Recovery and Reinvestment Act (ARRA) raised the national loan limit for Home Equity Conversion Mortgages (HECM) to $625,500 through December 31, 2009. In our survey of HECM lenders, we asked lenders about the influence the increased loan limit would have on their likelihood to offer HECMs and non-HECM reverse mortgages (see fig. 14). Additionally, we asked how they expected consumer demand for HECMs to increase as a result of the ARRA loan increase (see fig. 15). See figures 14 and 15 for survey questions and estimates based on our survey results.

Figure 14: Influence of ARRA’s Increase to Loan Limits on Lenders’ Plans to Offer Reverse Mortgages

Plans to offer HECMs

Plans to offer non-HECM reversemortgages 29 54 11 6 N/A

65% 34 % 0 % 1 % N/A %

Don’tknow




Notapplicable



Figure 15: Influence of ARRA’s Increase to Loan Limits on Consumer Demand for HECMs

Lenders’ expectation of ARRA’seffect on consumer demand for HECMs

82 % 13 % 0 % 5 % N/A %

Don’tknow

Increase somewhator greatly No effect

Decrease somewhat or greatly

Notapplicable



Reverse Mortgages

Appendix III: Effect of the Housing and

Economic Recovery Act of 2008 on Up-front

Costs for HECM Borrowers

Appendix III: Effect of the Housing and Economic Recovery Act of 2008 on Up-front Costs for HECM Borrowers

Home Equity Conversion Mortgage (HECM) borrowers may experience changes in up-front costs due to the Housing and Economic Recovery Act of 2008’s (HERA) change to the calculation of the origination fee, the loan limit, or both. Generally, borrowers with house values greater than the prior HECM loan limit will be able to borrow more under HERA’s higher loan limit, while borrowers with a wide range of house values may be affected by the changes in origination fees.1

There are two up-front costs. The first—the up-front mortgage insurance premium—is 2 percent of the maximum claim amount. The second—the origination fee—was calculated before HERA as 2 percent of the maximum claim amount with a minimum fee of $2,000. HERA changed the calculation of the origination fee to 2 percent of the first $200,000 of the maximum claim amount plus 1 percent of the maximum claim amount over $200,000, with a maximum fee of $6,000. In implementing HERA, HUD also increased the minimum origination fee by $500 to $2,500.

To determine how borrowers would be affected by these changes, we developed mathematical equations for calculating the up-front costs under both the HERA and pre-HERA rules. We subtracted the equation for the pre-HERA rules from the equation for the HERA rules to derive an equation for the change in up-front costs resulting from HERA. A positive value indicates that a borrower would pay more under HERA, and a negative value indicates that a borrower would pay less. Figures 16 and 17 illustrate how these changes affect different categories of borrowers.

Figure 16 shows the results for borrowers who have home values lower than the previous loan limit. The maximum claim amount is not affected by HERA’s change in loan limit. Therefore, for these borrowers, changes in up-front costs derive only from changes in the origination fee.

1HERA did not change the HECM loan limits in parts of Hawaii that had loan limits higher than $417,000 prior to HERA. As a result, borrowers in those areas whose house values are greater than the pre-HERA loan limits will see no change in their up-front mortgage insurance premiums due to HERA.


Appendix III: Effect of the Housing and

Economic Recovery Act of 2008 on Up-front

Costs for HECM Borrowers

Figure 16: Changes to Up-front Costs for Borrowers Not Affected by HERA’s Change in Loan Limit

Source: GAO.

Maximum claimamount (MC)

Equation for change in up-front costs due to HERA (dollars) Range of change to up-front costs

Borrow pays more

Borrower pays the same or less

$0 to $100,000

$100,000 to $125,000

$125,000 to $200,000

$200,000 to $400,000

$400,000 to $625,500

500

2,500 - (0.02 X MC)

0

2,000 - (0.01 X MC)

6,000 - (0.02 X MC)

$500 more

from $0 to 500 more

$0

from $0 to $2,000 less

from $2,000 to $6,510 less

Figure 17 shows the results of the calculation for borrowers who were affected by HERA’s increase in loan limit. These borrowers would pay up-front mortgage insurance premiums and origination fees based on a higher maximum claim amount. However, depending on the maximum claim amount, the origination fee may have decreased rather than increased. The net change in up-front costs for this grouping is therefore indeterminable without knowing the old and new maximum claim amounts.

Figure 17: Changes to Up-front Costs for Borrowers Affected by HERA’s Change in Loan Limit

Note: New maximum claim amounts and old maximum claim amounts less than $125,000 are not valid because no local loan limits were less than $125,000.


Appendix IV:

A

GAO Contact and Staff

cknowledgments

Page 48 GAO-09-836

Appendix IV: GAO Contact and Staff Acknowledgments

Mathew J. Scirè, at (202) 512-8678 or [email protected]

In addition to the individual named above, Steve Westley, Assistant Director; Anne Akin, Kathleen Boggs, Joanna Chan, Rudy Chatlos, Karen Jarzynka, John McGrail, Marc Molino, Mark Ramage, Carl Ramirez, Barbara Roesmann, and Jennifer Schwartz made key contributions to this report.

Reverse Mortgages

GAO Contact

Staff Acknowledgments

(250421)


GAO’s Mission The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO’s Web site (www.gao.gov). Each weekday afternoon, GAO posts on its Web site newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to www.gao.gov and select “E-mail Updates.”

Obtaining Copies of GAO Reports and Testimony

Order by Phone The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s Web site, http://www.gao.gov/ordering.htm.

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537.

Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information.

Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: [email protected] Automated answering system: (800) 424-5454 or (202) 512-7470

Ralph Dawn, Managing Director, [email protected], (202) 512-4400 U.S. Government Accountability Office, 441 G Street NW, Room 7125 Washington, DC 20548

To Report Fraud, Waste, and Abuse in Federal Programs

Congressional Relations

Chuck Young, Managing Director, [email protected], (202) 512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, DC 20548

Public Affairs

Please Print on Recycled Paper

http://www.gao.gov/

http://www.gao.gov/

http://www.gao.gov/ordering.htm

http://www.gao.gov/fraudnet/fraudnet.htm




GAO-09-836 Reverse Mortgages: Policy Changes …NRMLA National Reverse Mortgage Lenders Association...

Documents

Transcript of GAO-09-836 Reverse Mortgages: Policy Changes …NRMLA National Reverse Mortgage Lenders Association...